Analysis of relationship between mixed heavy metal exposure and early renal damage based on a weighted quantile sum regression and Bayesian kernel machine regression model

• Published in: Journal of trace elements in medicine and biology Vol. 84; p. 127438
• Main Authors: An, Qi, Wang, Qingyao, Liu, Rujie, Zhang, Jiachen, Li, Shuangjing, Shen, Weitong, Zhou, Han, Liang, Yufen, Li, Yang, Mu, Lina, Lei, Lijian
• Format: Journal Article
• Language: English
• Published: Germany Elsevier GmbH 01.07.2024
• Subjects: Bayesian kernel machine regression; Interactions; Mixed heavy metal exposure; Renal damage; Weighted quantile sum regression; Cd; Mn; Weighted quantile sum regression; SCr; UALB; CI; Co; EGFR; Cr; BKMR; As; Pb; VIF; PIP; UNAG; WQS; Bayesian kernel machine regression; Renal damage; ICP-MS; Ni; Interactions; CKD; Mixed heavy metal exposure; BMI
• ISSN: 0946-672X; 1878-3252
• DOI: 10.1016/j.jtemb.2024.127438

Occupation, environmental heavy metal exposure, and renal function impairment are closely related. The relationship between mixed metal exposure and chronic renal injury is inadequately described, and the interaction between each metal is poorly explored. This cross-sectional study assessed mixed heavy metal exposure in the general population and their relationship with early renal impairment, as well as possible interactions between metals. The study was conducted in two communities in Taiyuan City in northern China. Multiple linear regression, weighted quantile sum (WQS) and bayesian kernel machine regression (BKMR) regression were used to explore the relationship of mixed heavy metal exposure with indicators of early kidney injury (N-acetyl-β-D- glucosidase (UNAG), urinary albumin (UALB)). Meanwhile, BKMR was used to explore the possible interactions between mixed heavy metal and indicators of early kidney injury. Based on the WQS regression results, we observed adjusted WQS coefficient β (β-WQS) of 0.711 (95% CI: 0.543, 0.879). Notably, this change was primarily driven by As (35.6%) and Cd (22.5%). In the UALB model, the adjusted β-WQS was 0.657 (95% CI: 0.567, 0.747), with Ni (30.5%), Mn (22.1%), Cd (21.2%), and As (18.6%) exhibiting higher weights in the overall effect. The BKMR results showed a negative interaction between As and other metals in the UNAG and UALB models, a positive interaction between Mn and Ni and other metals. No significant pairwise interaction was observed in the association of metals with indicators of early kidney injury. Through multiple linear regression, WQS regression, and BKMR analyses, we found that exposure to mixed heavy metals such as Cd, Cr, Pb, Mn, As, Co and Ni was positively correlated with UNAG and UALB. Moreover, there are complex interactions between two or more heavy metals in more than one direction. [Display omitted] •Mixed heavy metal exposure is closely linked to renal function.•Mixed heavy metal exposure correlates positively with early renal function indicators.•Different heavy metals exhibit interaction effects in multiple directions.